# License:: MIT license
class Imdb
- IMDB = "http://us.imdb.com"
- TITLE_OR_NAME_MATCH = /<a href="(\/(?:title|name)\/(?:tt|nm)[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
- TITLE_MATCH = /<a href="(\/title\/tt[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
- NAME_MATCH = /<a href="(\/name\/nm[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
- CREDIT_NAME_MATCH = /#{NAME_MATCH}<\/td><td[^>]+> \.\.\. <\/td><td[^>]+>(.+?)<\/td>/
+ IMDB = "https://www.imdb.com"
+ TITLE_OR_NAME_MATCH = /<a\s+href="(\/(?:title|name)\/(?:tt|nm)[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ TITLE_MATCH = /<a\s+href="(\/title\/tt[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ NAME_MATCH = /<a\s+onclick="[^"]+"\s+href="(\/name\/nm[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ CHAR_MATCH = /<a\s+href="(\/character\/ch[0-9]+\/?)[^"]*"(?:[^>]*)>([^<]*)<\/a>/
+ CREDIT_NAME_MATCH = /#{NAME_MATCH}\s*<\/td>\s*<td[^>]+>\s*\.\.\.\s*<\/td>\s*<td[^>]+>\s*(.+?)\s*<\/td>/m
FINAL_ARTICLE_MATCH = /, ([A-Z]\S{0,2})$/
+ DESC_MATCH = /<meta name="description" content="(.*?)\. (.*?)\. (.*?)\."\s*\/>/
MATCHER = {
+ :character => CHAR_MATCH,
:title => TITLE_MATCH,
:name => NAME_MATCH,
:both => TITLE_OR_NAME_MATCH
end
def search(rawstr, rawopts={})
+ # allow the user to search directly for (movie) IDs
+ if rawstr.match /$tt\d+^/ then
+ return ["/movie/#{rawstr}/"]
+ end
str = CGI.escape(rawstr)
- str << ";site=aka" if @bot.config['imdb.aka']
+ str << "&site=aka" if @bot.config['imdb.aka']
opts = rawopts.dup
opts[:type] = :both unless opts[:type]
return do_search(str, opts)
matcher = MATCHER[opts[:type]]
if resp.code == "200"
+ if opts[:all]
+ return resp.body.scan(matcher).map { |m| m.first }.compact.uniq
+ end
m = []
m << matcher.match(resp.body) if @bot.config['imdb.popular']
if resp.body.match(/\(Exact Matches\)<\/b>/) and @bot.config['imdb.exact']
end
def grab_info(info, body)
- /<div class="info">\s+<h5>#{info}:<\/h5>\s+(.*?)<\/div>/mi.match(body)[1] rescue nil
+ /<div (?:id="\S+-info" )?class="(?:txt-block|see-more inline canwrap)">\s*<h[45](?: class="inline")?>\s*#{info}:\s*<\/h[45]>\s*(.*?)<\/div>/mi.match(body)[1] rescue nil
end
def fix_article(org_tit)
def info_title(sr, opts={})
resp = nil
begin
+ # movie urls without tailing / trigger a redirect
+ sr += '/' if sr[-1,1] != '/'
resp = @bot.httputil.get_response(IMDB + sr, :max_redir => -1)
rescue Exception => e
error e.message
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- title_date = m[1]
- pre_title, date, extra = title_date.scan(/^(.*)\((\d\d\d\d(?:\/[IV]+)?)\)\s*(.+)?$/).first
+ title_date = m[1].ircify_html
+ debug title_date
+ # note that the date dash for series is a - (ndash), not a - (minus sign)
+ # also, the second date, if missing, is an no-break space
+ pre_title, extra, date, junk = title_date.scan(/^(.*)\((.+?\s+)?(\d\d\d\d(?:–(?:\d\d\d\d| )?)?(?:\/[IV]+)?)\)\s*(.+)?$/).first
+ extra.strip! if extra
pre_title.strip!
- title = fix_article(pre_title.ircify_html)
+ title = fix_article(pre_title)
dir = nil
- data = grab_info(/Directors?/, resp.body)
+ data = grab_info(/(?:Director|Creator)s?/, resp.body)
if data
dir = data.scan(NAME_MATCH).map { |url, name|
name.ircify_html
country = data.ircify_html.gsub(' / ','/')
end
- info << [title, "(#{country}, #{date})", extra, dir ? "[#{dir}]" : nil, ": http://us.imdb.com#{sr}"].compact.join(" ")
+ info << [title, "(#{country}, #{date})", extra, dir ? "[#{dir}]" : nil, opts[:nourl] ? nil : ": http://www.imdb.com#{sr}"].compact.join(" ")
return info if opts[:title_only]
end
ratings = "no votes"
- m = /<b>([0-9.]+)\/10<\/b>\n?\r?\s+<small>\(<a href="ratings">([0-9,]+) votes?<\/a>\)<\/small>/.match(resp.body)
- if m
- ratings = "#{m[1]}/10 (#{m[2]} voters)"
+ # parse imdb rating value:
+ if resp.body.match(/itemprop="ratingValue">([^<]+)</)
+ ratings = "#{$1}/10"
+ end
+ # parse imdb rating count:
+ if resp.body.match(/itemprop="ratingCount">([^<]+)</)
+ ratings += " (#{$1} voters)"
end
genre = Array.new
- resp.body.scan(/<a href="\/Sections\/Genres\/[^\/]+\/">([^<]+)<\/a>/) do |gnr|
- genre << gnr
+ resp.body.scan(/<a\s+href="\/genre\/[^\?]+\?[^"]+"\s+>([^<]+)<\/a>/) do |gnr|
+ genre << gnr.first.strip
end
- plot = nil
- data = grab_info(/Plot (?:Outline|Summary)/, resp.body)
- if data
- plot = "Plot: " + data.ircify_html.gsub(/\s+more$/,'')
- end
+ plot = resp.body.match(DESC_MATCH)[3] rescue nil
+ # TODO option to extract the long storyline
+ # data = resp.body.match(/<h2>Storyline<\/h2>\s+/m).post_match.match(/<\/p>/).pre_match rescue nil
+ # if data
+ # data.sub!(/<em class="nobr">Written by.*$/m, '')
+ # plot = data.ircify_html.gsub(/\s+more\s*$/,'').gsub(/\s+Full summary » \| Full synopsis »\s*$/,'')
+ # end
+ plot = "Plot: #{plot}" if plot
info << ["Ratings: " << ratings, "Genre: " << genre.join('/') , plot].compact.join(". ")
if resp.code == "200"
m = /<title>([^<]*)<\/title>/.match(resp.body)
return nil if !m
- name = m[1]
+ name = m[1].sub(/ - IMDb/, '')
- info << "#{name} : http://us.imdb.com#{sr}"
+ info << name
+ info.last << " : http://www.imdb.com#{sr}" unless opts[:nourl]
return info if opts[:name_only]
}
end
- urls.map { |url|
- info = info_name(url, :movies_by_year => true)
+ urls.map { |u|
+ info = info_name(u, :movies_by_year => true)
debug info.inspect
end
def name_in_movie(name_urls, movie_urls)
+ debug name_urls
info = []
movie_urls.each { |movie|
title_info = info_title(movie, :title_only => true)
valid = []
data = @bot.httputil.get(IMDB + movie + "fullcredits")
- data.scan(CREDIT_NAME_MATCH).each { |url, name, role|
- valid << [url, name.ircify_html, role.ircify_html] if name_urls.include?(url)
+ data.scan(CREDIT_NAME_MATCH).each { |url, name, role_data|
+ ch_url, role = role_data.scan(CHAR_MATCH).first
+ debug [ch_url, role]
+ wanted = name_urls.include?(url) || name_urls.include?(ch_url)
+ valid << [url, name.ircify_html, role.ircify_html] if wanted
}
valid.each { |url, name, role|
info << "%s : %s was %s in %s" % [name, IMDB + url, role, title_info]
attr_reader :i
+ TITLE_URL = %r{^http://(?:[^.]+\.)?imdb.com(/title/tt\d+/)}
+ NAME_URL = %r{^http://(?:[^.]+\.)?imdb.com(/name/nm\d+/)}
+ def imdb_filter(s)
+ loc = Utils.check_location(s, TITLE_URL)
+ if loc
+ sr = loc.first.match(TITLE_URL)[1]
+ extra = $2 # nothign for the time being, could be fullcredits or whatever
+ res = i.info_title(sr, :nourl => true, :characters => (extra == 'fullcredits'))
+ debug res
+ if res
+ return {:title => res.first, :content => res.last}
+ else
+ return nil
+ end
+ end
+ loc = Utils.check_location(s, NAME_URL)
+ if loc
+ sr = loc.first.match(NAME_URL)[1]
+ extra = $2 # nothing for the time being, could be filmoyear or whatever
+ res = i.info_name(sr, :nourl => true, :movies_by_year => (extra == 'filmoyear'))
+ debug res
+ if res
+ name = res.shift
+ return {:title => name, :content => res.join(". ")}
+ else
+ return nil
+ end
+ end
+ return nil
+ end
+
def initialize
super
@i = Imdb.new(@bot)
+ @bot.register_filter(:imdb, :htmlinfo) { |s| imdb_filter(s) }
end
# Find a person or movie on IMDB. A :type (name/title, default both) can be
end
# Find the movies with a participation of :who in the year :year
- # TODO: allow year to be either a year or a decade ('[in the] 1960s')
+ # TODO: allow year to be either a year or a decade ('[in the] 1960s')
#
def movies(m, params)
who = params[:who].to_s
# Find the character played by :who in :movie
#
def character(m, params)
- who = params[:who].to_s
movie = params[:movie].to_s
-
- name_urls = i.search(who, :type => :name)
- unless name_urls
- m.reply "nothing found about #{who}, sorry"
+ movie_urls = i.search(movie, :type => :title)
+ unless movie_urls
+ m.reply "movie #{who} not found, sorry"
return
end
- movie_urls = i.search(movie, :type => :title)
- unless movie_urls
+ if params[:actor]
+ who = params[:actor].to_s
+ type = :name
+ else
+ who = params[:character].to_s
+ type = :character
+ end
+
+ name_urls = i.search(who, :type => type, :all => true)
+ unless name_urls
m.reply "nothing found about #{who}, sorry"
return
end
plugin.map "imdb [:type] *what", :requirements => { :type => /name|title/ }, :defaults => { :type => 'both' }
plugin.map "movies :prefix *who in [the] :years [as :role]", :requirements => { :prefix => /with|by|from/, :years => /'?\d+s?/ }
-plugin.map "character [played] by *who in *movie"
-plugin.map "character of *who in *movie"
+plugin.map "character [played] by *actor in *movie"
+plugin.map "character of *character in *movie"
plugin.map "characters in *movie", :action => :imdb
projects / user / henk / code / ruby / rbot.git / blobdiff